introduce jinja2 for preprocessing SQL

Replaces various hand-crafted replacements of varying format with
a single Jinja2 templating mechanism. Allows full access to
configuration if necessary.

nominatim/db/sql_preprocessor.py

@@ -0,0 +1,62 @@
+"""
+Preprocessing of SQL files.
+"""
+import jinja2
+
+
+def _get_partitions(conn):
+    """ Get the set of partitions currently in use.
+    """
+    with conn.cursor() as cur:
+        cur.execute('SELECT DISTINCT partition FROM country_name')
+        partitions = set([0])
+        for row in cur:
+            partitions.add(row[0])
+
+    return partitions
+
+
+def _get_tables(conn):
+    """ Return the set of tables currently in use.
+        Only includes non-partitioned
+    """
+    with conn.cursor() as cur:
+        cur.execute("SELECT tablename FROM pg_tables WHERE schemaname = 'public'")
+
+        return set((row[0] for row in list(cur)))
+
+class SQLPreprocessor: # pylint: disable=too-few-public-methods
+    """ A environment for preprocessing SQL files from the
+        lib-sql directory.
+
+        The preprocessor provides a number of default filters and variables.
+        The variables may be overwritten when rendering an SQL file.
+
+        The preprocessing is currently based on the jinja2 templating library
+        and follows its syntax.
+    """
+
+    def __init__(self, conn, config, sqllib_dir):
+        self.env = jinja2.Environment(autoescape=False,
+                                      loader=jinja2.FileSystemLoader(str(sqllib_dir)))
+
+        db_info = {}
+        db_info['partitions'] = _get_partitions(conn)
+        db_info['tables'] = _get_tables(conn)
+        db_info['reverse_only'] = 'search_name' not in db_info['tables']
+
+        self.env.globals['config'] = config
+        self.env.globals['db'] = db_info
+        self.env.globals['modulepath'] = config.DATABASE_MODULE_PATH or \
+                                         str((config.project_dir / 'module').resolve())
+
+
+    def run_sql_file(self, conn, name, **kwargs):
+        """ Execute the given SQL file on the connection. The keyword arguments
+            may supply additional parameters for preprocessing.
+        """
+        sql = self.env.get_template(name).render(**kwargs)
+
+        with conn.cursor() as cur:
+            cur.execute(sql)
+        conn.commit()

nominatim/tools/refresh.py

@@ -3,12 +3,12 @@ Functions for bringing auxiliary data in the database up-to-date.
 """
 import json
 import logging
-import re
 from textwrap import dedent
 
 from psycopg2.extras import execute_values
 
 from ..db.utils import execute_file
+from ..db.sql_preprocessor import SQLPreprocessor
 from ..version import NOMINATIM_VERSION
 
 LOG = logging.getLogger()
@@ -76,100 +76,17 @@ def load_address_levels_from_file(conn, config_file):
     with config_file.open('r') as fdesc:
         load_address_levels(conn, 'address_levels', json.load(fdesc))
 
-PLPGSQL_BASE_MODULES = (
-    'utils.sql',
-    'normalization.sql',
-    'ranking.sql',
-    'importance.sql',
-    'address_lookup.sql',
-    'interpolation.sql'
-)
 
-PLPGSQL_TABLE_MODULES = (
-    ('place', 'place_triggers.sql'),
-    ('placex', 'placex_triggers.sql'),
-    ('location_postcode', 'postcode_triggers.sql')
-)
-
-def _get_standard_function_sql(conn, config, sql_dir, enable_diff_updates, enable_debug):
-    """ Read all applicable SQLs containing PL/pgSQL functions, replace
-        placefolders and execute them.
-    """
-    sql_func_dir = sql_dir / 'functions'
-    sql = ''
-
-    # Get the basic set of functions that is always imported.
-    for sql_file in PLPGSQL_BASE_MODULES:
-        with (sql_func_dir / sql_file).open('r') as fdesc:
-            sql += fdesc.read()
-
-    # Some files require the presence of a certain table
-    for table, fname in PLPGSQL_TABLE_MODULES:
-        if conn.table_exists(table):
-            with (sql_func_dir / fname).open('r') as fdesc:
-                sql += fdesc.read()
-
-    # Replace placeholders.
-    sql = sql.replace('{modulepath}',
-                      config.DATABASE_MODULE_PATH or str((config.project_dir / 'module').resolve()))
-
-    if enable_diff_updates:
-        sql = sql.replace('RETURN NEW; -- %DIFFUPDATES%', '--')
-
-    if enable_debug:
-        sql = sql.replace('--DEBUG:', '')
-
-    if config.get_bool('LIMIT_REINDEXING'):
-        sql = sql.replace('--LIMIT INDEXING:', '')
-
-    if not config.get_bool('USE_US_TIGER_DATA'):
-        sql = sql.replace('-- %NOTIGERDATA% ', '')
-
-    if not config.get_bool('USE_AUX_LOCATION_DATA'):
-        sql = sql.replace('-- %NOAUXDATA% ', '')
-
-    reverse_only = 'false' if conn.table_exists('search_name') else 'true'
-
-    return sql.replace('%REVERSE-ONLY%', reverse_only)
-
-
-def replace_partition_string(sql, partitions):
-    """ Replace a partition template with the actual partition code.
-    """
-    for match in re.findall('^-- start(.*?)^-- end', sql, re.M | re.S):
-        repl = ''
-        for part in partitions:
-            repl += match.replace('-partition-', str(part))
-        sql = sql.replace(match, repl)
-
-    return sql
-
-def _get_partition_function_sql(conn, sql_dir):
-    """ Create functions that work on partition tables.
-    """
-    with conn.cursor() as cur:
-        cur.execute('SELECT distinct partition FROM country_name')
-        partitions = set([0])
-        for row in cur:
-            partitions.add(row[0])
-
-    with (sql_dir / 'partition-functions.src.sql').open('r') as fdesc:
-        sql = fdesc.read()
-
-    return replace_partition_string(sql, sorted(partitions))
-
-def create_functions(conn, config, sql_dir,
+def create_functions(conn, config, sqllib_dir,
                      enable_diff_updates=True, enable_debug=False):
     """ (Re)create the PL/pgSQL functions.
     """
-    sql = _get_standard_function_sql(conn, config, sql_dir,
-                                     enable_diff_updates, enable_debug)
-    sql += _get_partition_function_sql(conn, sql_dir)
+    sql = SQLPreprocessor(conn, config, sqllib_dir)
 
-    with conn.cursor() as cur:
-        cur.execute(sql)
+    sql.run_sql_file(conn, 'functions.sql',
+                     disable_diff_update=not enable_diff_updates,
+                     debug=enable_debug)
 
-    conn.commit()
 
 
 WEBSITE_SCRIPTS = (